Levalbuterol vs. albuterol for hospitalized patients with COPD in China: cost-utility and budget impact analysis

Abstract

Objectives

For hospitalized patients with chronic obstructive pulmonary disease (COPD), albuterol and levalbuterol can both be used as relievers to alleviate bronchoconstriction. This study aimed to evaluate levalbuterol and albuterol's cost-utility and budget impact in hospitalized patients with COPD.

Interventions

A cost-utility analysis was used to evaluate the impact on the costs of nebulized levalbuterol verse albuterol in hospitalized patients with COPD. The decision tree model was employed to estimate the incremental cost per quality-adjusted life year in the admission setting. A budget impact model was used to examine the impact of budget on levalbuterol's entry into the Chinese market from the healthcare system's perspective. One-way sensitivity and probabilistic sensitivity analyses were performed to test the uncertainty of the parameters.

Results

The cost-utility results showed that levalbuterol saved ¥495.7 ($105.1) per hospitalization, while the budget impact analysis revealed a potential saving of ¥22.3 ($6.8) million in 3 years. The sensitivity analysis indicated that the results were robust to the changes in input parameter values.

Conclusion

Levalbuterol is a cost-saving option for treating hospitalized patients with COPD in China.

PLAIN LANGUAGE SUMMARY

Chronic obstructive pulmonary disease (COPD) is a common disease in China, with an increased financial burden over the years. Nebulized albuterol is the most commonly used short-acting beta2-agonist, often regarded as the initial bronchodilator to treat hospitalized COPD patients. Its R-isomer, levalbuterol, entered the Chinese market in 2019. The new intervention always impacts the expenditure of the health system. We built a cost-utility and budget impact model to analyze the difference between albuterol and levalbuterol. The cost-utility results showed that levalbuterol saved ¥495.7 ($105.1) per hospitalization compared with albuterol, while the budget impact analysis revealed a potential saving of ¥22.3 ($6.8) million in 3 years.

Keywords:

• Albuterol
• budget impact analysis
• chronic obstructive pulmonary disease
• cost-utility analysis
• levalbuterol

JEL CLASSIFICATION CODES:

• A13
• A1
• A
• A10

Introduction

Chronic obstructive pulmonary disease (COPD) is a common disease in China¹, with an increased financial burden over the years². The direct medical cost of COPD ranges from 72 to 3,565 United States dollars per capita per year in China, accounting for 33.33–118.09% of local average annual income^3,4. Medicine cost was the biggest contributor (45.53%)⁵. Considering the financial burden, cost-effective drugs should be used in COPD treatments.

Nebulized inhaled short-acting beta2-agonists (SABA) are considered an easier delivery method for sicker patients and are recommended as the initial bronchodilators to treat hospitalized patients with COPD^6,7, and the most commonly used SABA is albuterol. Levalbuterol is the R-isomer of albuterol⁸, which entered the Chinese market in 2019⁹. Compared with albuterol, nebulized levalbuterol could significantly improve lung symptoms in patients with COPD and reduce the risk of adverse events and exacerbation^10,11, resulting in 1 day less hospitalization and $553 saving per patient with COPD over the exacerbation period¹². However, this pattern of results is not universal^11,13,14. These conflicting results have made it difficult for physicians to choose better nebulized SABAs for treating patients with COPD.

This study aimed to perform a cost-utility analysis (CUA) comparing nebulized levalbuterol with albuterol in hospitalized patients with COPD from a Chinese societal perspective. A budget impact analysis (BIA) was also conducted to evaluate the impact on the Chinese healthcare budget of levalbuterol for treating COPD over three years.

Methods

Cost-utility analysis

Cost-utility model, perspective, and model outcomes

The assessment of the cost-difference and health outcomes of the use of levalbuterol vs. albuterol was performed using a decision tree model from a Chinese societal perspective. The decision tree model was built based on China Guidelines for Pharmacoeconomic Evaluations¹⁵, to estimate the incremental cost per quality-adjusted life year in one admission setting. Since as need use SABAs are the initial bronchodilators to treat hospitalized patients with COPD⁷, which don’t have an impact on the prognosis of disease status, we only considered the short-term economic differences between levalbuterol and albuterol during hospitalization. Microsoft Excel 2016 software was used to analyze the model. The model structure was presented in Supplemental Figure 1. According to guidelines and previously published studies^15–17, health outcomes were presented as an additional unit of utility (QALY). QALY is widely used for comprehensively reflecting treatment effects, such as clinical efficacy, patient psychology, and living status. If the levalbuterol has a lower cost and better outcome than the albuterol, it will be strictly dominant; if the levalbuterol has both a higher cost and a better outcome than the albuterol, the incremental cost-utility ratio (ICUR) needs to be calculated¹⁵. The ICUR was associated with cost and QALY, and was calculated using the following formula. The target population was hospitalized patients with COPD. $$\text{ICUR}=\left({\text{Cost}}_{\text{levalbuterol}}–{\text{Cost}}_{\text{albuterol}}\right)/\left({\text{QALY lost}}_{\text{levalbuterol}}–{\text{QALY lost}}_{\text{albuterol}}\right)$$

Intervention, assumption, and time horizon

Considering the as-needed usage for nebulized SABAs with no strict determined dosage, the maximum doses were chosen according to package inserts [levalbuterol, 1.25 mg, three times a day (tid) vs. albuterol, 5 mg, four times a day (qid)]. A 1-year time horizon was adopted to capture the short-term clinical and economic outcomes in patients with COPD, and thus discounting was not applied¹⁵. The model applied the following assumptions: (A) The frequency of COPD-related hospitalization was assumed to be once a year per patient, and the treatments of levalbuterol and albuterol continued throughout the hospitalization so that the length of hospital stay was equal to the administration time. All patients were treated with the same basic and maintenance therapy. (B) No people died in the study.

Base case model parameters

To compare the differences in efficacy, safety, and economics between levalbuterol and albuterol, a structured literature review was conducted on PubMed, Cochrane Library, Medline, Embase, and Web of Science from the earliest record date to June 2021. In total, 66 studies were identified and individually screened. The final included studies were based on the following criteria:

1. Clinical trials

2. Hospitalized COPD patients

3. Comparing levalbuterol with albuterol, nebulized administration

4. The article is written in English or Chinese

5. Outcomes with the length of stay per hospitalization

The information obtained through the literature search was used in the cost-utility model (implemented in Microsoft Excel). The parameters used in the model are summarized in Table 1.

Table 1. Summary of cost-utility model inputs.

Parameters	Value	Source
Efficacy data
Levalbuterol: length of hospitalization (day)	5.1	Truitt et al.^12
Albuterol: length of hospitalization (day)	6.1	Truitt et al.^12
Utility
Well-controlled state	0.989	Rodriguez-Martinez et al.^18
Severe COPD state (hospitalization)	0.751	Hertel et al.^19
Duration of a single attack (h)	1	Experts' investigation
Numbers of daily nebulization (numbers of daily attacks)
Levalbuterol	3	Package insert
Albuterol	4	Package insert
Costs
Levalbuterol, per 1.25 mg (unit price)	¥29.12	Provincial/municipal drug procurement databases^20
The weighted average price of albuterol, per 5 mg (unit price)	¥21.62
Nursing costs for nebulization, per event	¥7	Compilation of medical service items and price in Chengdu^21
Nursing expenses, per day	¥10
Intravenous infusion, per day	¥18
Oxygen therapy, per day	¥30
Ward costs, per day	¥60
Productivity costs, per day	¥116.05	Calculated based on Chinese GDP, Statistical Communique of the People's Republic of China on the 2019 National Economic and Social Development^22

Clinical data

Due to the limited published evidence, the length of hospitalization was the only efficacy parameter that could be used to build the model. Clinical trials did not report other factors associated with hospitalization costs, such as comorbidity, surgery, and antibiotic use.

The mean length of hospital stay in the levalbuterol group was one day less than that in the albuterol group (5.1 vs. 6.1 days)¹². Adverse drug reactions (ADRs) were not considered because of varied results¹⁰ and the low incidence of ADRs in inhaled SABAs.

Utilities

Health-related quality-of-life inputs (utilities) were applied to each health state to generate QALYs. Only two studies reported utilities for well-controlled and severe COPD states after the literature search. One study calculated the utility of 0.751 for severe COPD state based on European Quality of Life 5 Dimensions (EQ-5D)¹⁹, while the other study reported the utility of 0.989 for “no symptoms” asthma¹⁸, based on a utility valuation survey with standard gamble methodology. As the result of the literature search and recommendation of COPD guidelines^6,7 and experts' investigation, we assumed the utility values for the “hospitalization” state and “well-controlled COPD” state were 0.751 and 0.989, respectively. Based on the clinical practice, drugs were administered right after the attack. The duration of single attacks to drug alleviation was set to 1 h, and the drug administration times were regarded as the number of attacks. According to the label inserts, the usage of levalbuterol was tid, with qid for albuterol. The QALY lost in COPD was calculated using the following formula: $$\begin{array}{c}\text{QALY lost}=\left(\text{well}-\text{controlled state utility}-\text{severe COPD state utility}\right)\\ \times \text{duration of single attack}\times \text{numbers of daily attacks}\times \text{length of hospital stay}\end{array}$$

Costs

The analysis considered both direct and indirect medical costs from the societal perspective. Direct medical costs comprised drug costs, nursing costs, and hospitalization costs. Drug costs were calculated based on the average wholesale prices and market shares in 2020 and are presented in Supplemental Table 2. Nursing costs included costs of nebulization (¥7 per event) and nursing (¥10 per day), while hospitalization costs included intravenous infusion (¥18 per day), oxygen therapy (¥30 per day), and ward costs (¥60 per day)²³. The human capital approach was used to calculate the indirect medical costs¹⁵, which consisted of family members or guardians' productivity loss. Considering the existence of patient's own productivity, it was necessary to calculate the labor loss of the patient and the caregiver¹⁵. The costs related to drug and hospitalization (such as intravenous infusion and oxygen therapy costs) are presented in Table 1. Some key points of price (CNY) were included USD using the average exchange rate for the year 2022²⁴.

Scenario analysis and sensitivity analysis

Three scenarios were considered. In scenario 1, considering the conflicting conclusion in levalbuterol benefits compared with albuterol, we assumed that the length of hospital stay was similar between the two SABAs. Thus, the length of stay in the two groups was 6.1 days. In scenario 2, we only considered the direct costs, such as drug, nursing, and hospitalization costs, but no indirect costs like productive costs. Moreover, scenario 3 neglected the duration differences between levalbuterol and albuterol, which resulted in the same number of daily nebulization and attacks.

One-way sensitivity analysis was performed to explore uncertainty in individual parameters. Inputs were varied to their upper and lower 95% CIs to determine which factors impacted the model most. A probabilistic sensitivity analysis (PSA) was also undertaken to assess the robustness of the ICUR to simultaneous changes in parameters, in which most parameters were input not as fixed values but as distributions. Distributions were chosen based on data from the literature or on plausible assumptions and experts' opinions when no clear evidence was available. The ICUR obtained from 1,000 iterations and stochastically sampled from the distributions were used to estimate a mean value. The cost-utility acceptability curve determined the probability of the ICUR being less than the willingness-to-pay (WTP) value. The threshold of three times annual gross domestic product (GDP) per capita (¥212,676, $64,388.6)²² was applied since the WTP per QALY is not yet determined in China^15,25,26. The range of input parameters in the sensitivity analysis is summarized in Supplemental Table 1.

Budget impact analysis

Budget impact model, perspective, and time horizon

Based on the conclusion from the CUA model, a companion budget impact model (BIM) was developed to address the hypothetical changes to the Chinese Healthcare Service budget considering the market expansion of levalbuterol from the healthcare system perspective. The model structure was presented in Supplemental Figure 2. The BIA for hospitalized patients was based on epidemiological data and expert investigation over a 3-year time horizon (2021–2023).

Target population, intervention, and market shares

Chinese epidemiologic and demographic data used to calculate the BIM target population are included in Table 2. The intervention was the same as the CUA.

Table 2. Summary of budget impact model inputs.

Parameters	Value	Source
Incidence
Gross population in 2019	1,400,050,000	Statistical Communique of the People's Republic of China on the 2019 National Economic and Social Development^22
Population growth in 2019	0.46%
The proportion of the adult over-40 population	42.60%	Tabulation on the 2010 population census of the People's Republic of China^27
The prevalence rate of COPD	13.70%	Wang et al.^1
Diagnostic rate of COPD	15.34%	Zhong et al.^28
Hospitalization rate	3.00%	Fang et al.^29
The proportion of patients treated with nebulized SABAs	37.8%	Singh et al.^30
Nationwide healthcare insurance coverage	96.74%	Statistical Communique of the People's Republic of China on the 2019 National Economic and Social Development^22
Costs
Levalbuterol: hospitalization costs per visit (5.1 days)	¥10,014.8	Yang^31
Albuterol: hospitalization costs per visit (6.1 days)	¥11,978.5

Market share data were obtained from China Hospital Pharmaceutical Audit in 2020. Current market share projections of SABAs from the years 2021 through 2023 were used for the “Before levalbuterol entry National Reimbursement Drug List (NDRL)” and then adjusted for expected change for “levalbuterol post-entry.” Annual costs associated with admission setting were estimated based on the yearly cost of each treatment and the projected market share of each treatment. The detailed market share is shown in Supplemental Tables 2 and 3.

Resource used and costs in the base case

From the healthcare system perspective, only direct medical costs were considered, for the indirect costs are usually irrelevant to decision-makers. Considering the as-needed use of SABAs, the length of drug therapy was equivalent to the length of hospital stay. Inputs related to hospitalization costs applied for the BIM were presented in Table 2.

Scenario and sensitivity analyses

Potential arguments may lead to varied conclusions on levalbuterol benefits. Hence, a no-benefit scenario analysis was conducted with the same length of stay in two groups (6.1 days). A univariate sensitivity analysis was conducted to test the robustness of the results by modifying the following parameters: prevalence, diagnosis rate, rate of hospitalization for COPD, market share of levalbuterol, length of hospitalization, and price of levalbuterol. The range of input parameters in the sensitivity analysis is summarized in Supplemental Table 4.

Patient and public involvement

The study did not involve any direct patient and public involvement.

Results

Cost-utility analysis

In the base case, levalbuterol saved ¥495.7 ($150.1) per hospitalized patient compared with albuterol. The results are shown in Table 3. Improved cost-saving and QALY were established owing to the reduced length of hospital stay in the levalbuterol group. In the scenario analysis, each patient could save ¥37.2 ($11.3), ¥263.6 ($79.8), and ¥311.5 ($94.3) in scenario 1, scenario 2, and scenario 3, separately.

Table 3. CUA results.

Comparator	Drug costs (¥)	Nursing costs (¥)	Hospitalization costs (¥)	Indirect costs (¥)	Total Cost (¥)	Incremental QALY
Base case
Levalbuterol	445.5	158.1	550.8	1,183.7	2,338.2	0.0002483
Albuterol	527.4	231.8	658.8	1,415.8	2,833.9
Scenario 1
Levalbuterol	532.9	189.1	658.8	1,415.8	2,796.6	0.0001664
Albuterol	527.4	231.8	658.8	1,415.8	2,833.9
Scenario 2
Levalbuterol	445.5	158.1	550.8	0	1,154.4	0.0002483
Albuterol	527.5	231.8	658.8	0	1,418.0
Scenario 3
Levalbuterol	594.0	193.8	550.8	1,183.7	2,522.4	0.0001091
Albuterol	527.5	231.8	658.8	1,415.8	2,833.9

Abbreviations. QALY: quality-adjusted life year.

Sensitivity analysis

The results of the one-way sensitivity analysis are shown in a tornado diagram (Figure 1(a)). The diagram shows the effects of uncertainty on the ICUR for each parameter. The utility value of well-controlled and severe COPD states had the most significant impact on the model, while the lengths of stay in the two regimens and the weighted average price of albuterol also play crucial roles. The ICUR resulting from each PSA iteration is shown in Figure 1(b). Points lying below the frontier represent ICUR at the conventional cost/QALY threshold of ¥212,676 ($64,388.6). No inputs changed the conclusion, which supported the robustness of the model.

Figure 1. Sensitivity analysis. (a) Tornado diagram of univariate (one-way) sensitivity analysis. (b) ICUR scatterplot for levalbuterol vs. albuterol. This graph represents the incremental cost and corresponding incremental effect for each iteration of the probabilistic sensitivity analysis. The central dot represents the base case analysis ICUR.

Budget impact analysis

From the Chinese healthcare system perspective, the target number of hospitalized patients with COPD in 2020 was estimated to be 138,132 in China. Hospitalized patients with COPD in this group were eligible for levalbuterol vs. albuterol.

Levalbuterol led to a total budget saving of ¥ 22.3 ($6.8) million after three years of entry on the market. The total budget-saving included a decrease in drug costs (¥ 0.9, $0.3 million) and hospitalization (¥ 21.4, $6.5 million).

Scenario and sensitivity analyses

A scenario analysis was conducted considering no benefits in the decreased length of stay. Despite the unconsidered benefit, no significant impact of levalbuterol entry on the Chinese healthcare budget was observed, with a slight increase in drug costs in three years (¥7,401, $2,240.7). Given the uncertainty of all the parameters for the two treatments, a sensitivity analysis of the base case was conducted using plausible ranges given in Supplemental Table 4. Prevalence rate, diagnosis rate, and length of stay in the two regimes were found to have the highest impact on the model results, while the nebulized SABAs, hospitalization rate, and market share of levalbuterol were the underlying drivers of the results in the model (Figure 2).

Figure 2. One-way sensitivity results in budget impact analysis.

Discussion

The results from both CUA and BIA indicated that levalbuterol could be considered a cost-effective therapy in hospitalized patients with COPD compared with albuterol. The analysis suggested that using levalbuterol resulted in lower costs and a gain in QALYs. The total medical costs were ¥495.7 ($150.1) less than albuterol, with an 0.0002483 QALY gained per patient in the levalbuterol group. Regarding the albuterol alternative, levalbuterol could save ¥22.3 ($6.8) million in hospitalized patients with COPD over three years from the Chinese healthcare system perspective.

Concerning efficacy, some investigators found that using levalbuterol resulted in fewer hospital stays and exacerbations^10,12 and less need for rescue treatments¹⁰ in patients with COPD. However, the pattern of results was quite controversial, and some studies suggested opposite opinions^11,13,32. Due to limited published evidence comparing the efficacy of levalbuterol and albuterol, the study by Truitt et al.¹² was the only one showing a reduction in hospital stay, which might mainly be due to clinical benefits, such as better lung function and symptoms. Our result was consistent with published studies in the US, where levalbuterol showed a significant saving of ∼$967 for patients with COPD³³, indicating that cost-effectiveness derived mainly from hospitalization³⁴. The no-benefit situation was discussed in scenario 1 analysis, which was almost consistent with the base case.

The present cost-utility and budget impact models involved conservative estimations for the following reasons. First, without considering the benefits of ADRs and direct non-medical costs (such as nutrition costs), the present study might underestimate the cost-saving of levalbuterol. A randomized, multi-center, controlled trial reported that patients in the levalbuterol treatment group experienced significantly fewer β-mediated adverse effects (which might be caused by S-isomer) than patients in the albuterol treatment group (e.g. rapid heartbeat, 17.3 vs. 6.1%)¹¹. However, most were mild or moderate in severity. Considering the very low probability of severe ARDs of inhaled SABA, we assumed that levalbuterol and albuterol had the same safety profile. Second, China's rate of hospitalization due to COPD might be underestimated because of the lack of recent epidemiological studies. Zhang et al. reported that the inpatient rate would be 36.4%³⁵, while Fang et al. suggested the rate as 3.0%²⁹. This vast gap might derive from the sample size, geographic region, hospital grade, and data resources. The nationwide, cross-sectional study reported that the rate of hospitalized patients with COPD in China was 3% and was elected to build the BIM. However, this cross-sectional study was conducted in 2015. The prevalence and diagnostic rates might increase over the past seven years due to the enhancement of cognition in COPD³, which might result in population underestimation.

However, this study had several limitations. First, the low number of data resources might have reduced the precision of the results. For example, the well-controlled state's utility was derived from a CUA of asthma instead of COPD conducted in Colombia¹⁸. The utility of “no symptoms” asthma might over-evaluate the utility value of a well-controlled state in COPD patients might under-evaluate the benefits of levalbuterol. However, the main symptoms of asthma and COPD were respiratory symptoms and airflow limitation^6,36. SABAs were the standard rescue therapy for both diseases; therefore, the shared utility in the well-controlled state may be acceptable. Second, the generalizability of the results might be limited. Since the obtained outcomes have not been proven in the Chinese population, it will be necessary to explore the related parameters in China in future studies. Once obtain such a parameter, we will consider updating it. Also, the numeric parameters were not universal for all patients. We assumed that the hospitalization visit was set to be once a year in the QALYs calculation and BIM, but it would be ranged depending on individual characteristics³⁷. Meanwhile, the reimbursement percentage varied among regions since 65% was inputted. The model could better reflect the actual cost and budget-saving in each region and facilitate region-wise decision-making when adjusted with region-specific values and real-world data.

Conclusions

Overall, the analysis was based on varying evidence and assumptions, so the results need to be considered cautiously. The study combined a CUA with BIA, and the findings support the use of levalbuterol in hospitalized patients with COPD.

Transparency

Declaration of funding

This research received no specific grant from any funding agency in public, commercial, or not-for-profit sectors.

Declaration of financial/other relationships

The authors report that there are no competing interests to declare.

Author contributions

LC and XC participated in the study's design, adapted the model, performed the economic analysis, and drafted the manuscript. CZ and XL participated in interpreting the results and model adaptation and critically revised the article. YH helped and oversaw the study, reviewed the manuscript, and acted as a guarantor for the overall content. All authors proofread and approved the final manuscript and agree to be accountable for all aspects of the work.

Acknowledgements

The authors thank Dandan Zhao (Joincare, Shenzhen, China) for editorial support during the development of this manuscript.

Data availability statement

The datasets generated and analyzed during the current study are available from the corresponding author on reasonable request.

Reviewer disclosures

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.